DSXchange

Hi

I have a requirement like the datastage job should automate to run for every 5 mins.
first time the job is run and file loaded in /tmp/target/ and second time when it run the new file should check the key value on the existed data (that was run before) and if no key value exists then the record should CREATE (insert in the target file) and if the key exist in the target file then the values should UPDATE.

Please could any one help me to approach this requirement.

Regards,
Cheru.

It looks like, do you want to achieve insert/update mechanism in sequential file?

Start by posting in the correct forum.

pandeesh wrote:It looks like, do you want to achieve insert/update mechanism in sequential file?

YES its an INSERT/UPDATE mechanism in file.. please explain me.. thank you.

There is no mechanism to "update" a sequential file. You'll need to compare both sources and either write out new/updated records from the new source or any original records not 'updated'.

I believe you can use change capture stage for this.

pandeesh wrote:I believe you can use change capture stage for this.

---

I tried using change capture stage and it worked as expected

Thank you for the help..

chulett wrote:There is no mechanism to "update" a sequential file.

Are you sure?

Quite. That's the nature of sequential media. Now, we're not talking via an editor, stream or not, I'm saying you can't treat it like a relational source and update a specific record in place... unless we've moved into the realm of indexed sequential files.

chulett wrote:you can't treat it like a relational source

Quite certain are you? What know you of certainty?

Away put your weapon.

Setup the ODBC Text driver you can.

That path leads to the Dark Side.

Craig... I am your father.

Hi Eric,

I did not review the work done by my team on using the Change capture stage..

but told its worked .. let me check the same and will revert if its worked or not.

Regards,
Cheru.